Dev & Coding

A hard-won playbook for adversarially testing the fdeops CLI's engagement resolution, registry, hooks and <private> redaction boundary inside a throwaway sandbox.

A reference skill that teaches Claude the correct Python API for searching, renting, and managing GPUs on Vast.ai.

Give it a PR number and it checks out the branch, rebases onto main, triages every open review thread, and validates changed skills.

A quiet prompt mentor that scores each prompt by stage and appends one coaching line to every answer in Claude Code or Codex.

A five-mode testing workflow that finds high-value test candidates, writes and reviews focused tests, generates adversarial workloads, and diagnoses test-suite health.

Battle-tested configuration patterns for running Playwright tests fast and reliably on GitHub Actions, GitLab CI, Docker, and other CI providers.

Scaffolds a React + Tailwind + shadcn/ui project and bundles it into a single self-contained HTML artifact for claude.ai.

A five-gate Socratic workflow that drives a one-line idea into an approved spec with single-test acceptance criteria before any code is written.

An on-demand audit that finds provenance-tracked docs gone stale from out-of-commit source changes and lets you re-scout, re-baseline, or defer each one.

Delegate coding tasks to Mistral Vibe CLI while Claude decomposes, supervises via git diff, and reports token cost.

Gives Claude Code grounded, cited recall over the memory, rules and decisions that every coding agent on the machine has written, all stored locally.

An end-to-end pipeline that rewrites and re-verifies an existing Lattice skill to modern-model precision with provably zero behavioral loss.

Audits a skill through three independent reviewer personas and reports only the high-confidence, real-world gaps with concrete fixes, ordered by severity.

Treats the Lattice repo's skills/ directory as the source of truth and audits then fixes documentation drift across README, docs/, PROJECT.md, CLAUDE.md and GitHub issue templates.

An advanced playbook for designing or refactoring agent-based engineering workflows with bounded autonomy, custom evals, durable state, and verified write-back.

Turns an experiment blueprint JSON into a runnable Python project skeleton with baselines, training, evaluation and ablation configs.

A pattern library for setting up reliable, scalable scheduled functions (cron jobs) in Convex apps.

Ready-to-use Convex patterns for file uploads, URL serving, deletion, and `_storage` metadata access.

Builds or refreshes the Operator Project Index by driving the `operator-helper` CLI end to end.

A code-pattern skill for building stateful AI agents on Convex with threads, tools, streaming, RAG and multi-step workflows.